The planned study in this research grant application addresses the problem of growing and unacceptably high rates of drug use among adolescent girls in the United States. We propose to develop and test an interactive, web-based program to prevent drug abuse and delay the onset of drug use among girls. Study participants will be 13- and 14-year-old girls recruited from Facebook. Facebook is the largest and most popular social networking site in the United States, with an enrollment of 44 percent of all 13- and 14-year-old girls. Grounded in social cognitive theory and within a resiliency framework, the prevention program will equip girls with cognitive-behavioral skills to address peer and social influences, self-esteem, depression, and body- image issues associated with drug use among adolescent girls. All girls will access measures online. Girls in the intervention arm will access the program's initial sessions and annual booster sessions online. The gender- specific content will be housed as a menu of sessions embedded on a website offering girls access to current events, entertainment news, horoscopes, inspirational quotes, health and beauty tips, and other age- and gender-relevant features. Through virtual role-play simulations, the sessions will employ emotionally responsive avatars and a branched learning environment through which girls can acquire, try out, receive feedback on, and master new cognitive-behavioral skills. Interactive and gender-specific prevention content will give the program thematic coherence and enhance its attraction and value to adolescent girls. The study design is a randomized clinical trial. Recruited from across the United States, 13- and 14-year- old girls will be enrolled when they assent to study participation and obtain parental permission. Online, all girls will complete pretest measures; intervention-arm girls will interact with the prevention program, and all girls will complete posttest measures and 1-, 2-, and 3-year follow-up measures. Intervention-arm girls will additionally receive booster sessions after they complete 1- and 2-year follow-ups. Across follow-up measurement occasions, data analyses will examine rates of past 30-day drug use and drug use initiation between the intervention and control arms. We will also examine intervention effects on mediator variables associated with drug use risk and protective factors. And, we will assess the extent to which changes in mediator variables explain differences in drug use between arms.